Crustal structure beneath the Rif Cordillera, North Morocco, from the RIFSIS wide-angle reflection seismic experiment

Alba Gil1, Josep Gallart1, Jordi Diaz1, Ramon Carbonell1, Montserrat Torne1,

Alan Levander2, Mimoun Harnafi3

1. Earth Structure and Dynamics, Institute of Earth Sciences Jaume Almera-CSIC, Barcelona

Spain

2. Earth Sciences, Rice University, Houston, TX, United States

3. Institut Scientifique, Universit Mohammed V-Agdal, Rabat, Morocco

Abstract

The different geodynamic models proposed since the late 90's to account for the

complex evolution of the Gibraltar Arc System lack definite constraints on the crustal

structure of the Rif orogen. Here we present the first well-resolved P-wave velocity

crustal models of the Rif cordillera and its southern continuation towards the Atlas

made using controlled-source seismic data. Two 300+ km-long wide-angle reflection

profiles crossed the Rif along NS and EW trends. The profiles recorded simultaneously

5 land explosions of 1Tn each using ~850 high frequency seismometers. The crustal

structure revealed from 2D forward modeling delineates a complex, laterally-varying

crustal structure below the Rif domains. The most surprising feature, seen on both

profiles, is a ~50 km deep crustal root localized beneath the External Rif. To the east,

the crust thins rapidly by 20 km across the Nekkor fault, indicating that the fault is a

crustal scale feature. On the NS profile the crust thins more gradually to 40 km

thickness beneath Middle Atlas and 42 km beneath the Betics. These new seismic

results are in overall agreement with regional trends of Bouguer gravity and are

consistent with recent receiver function estimates of crustal thickness. The complex

crustal structure of the Rif orogen in the Gibraltar Arc is a consequence of the Miocene

Research Article Geochemistry, Geophysics, GeosystemsDOI 10.1002/2014GC005485

This article has been accepted for publication and undergone full peer review but has not beenthrough the copyediting, typesetting, pagination and proofreading process which may lead todifferences between this version and the Version of Record. Please cite this article asdoi: 10.1002/2014GC005485

2014 American Geophysical UnionReceived: Jul 07, 2014; Revised: Oct 10, 2014; Accepted: Nov 06, 2014

2

collision between the Iberian and African plates. Both the abrupt change in crustal

thickness at the Nekkor fault and the unexpectedly deep Rif crustal root can be

attributed to interaction of the subducting Alboran slab with the North African passive

margin at late Oligocene-early Miocene times.

Keywords: Seismic profiling, Western Mediterranean, Rif Cordillera, Crustal structure,

Moho depth variations

1. Introduction

At the westernmost Mediterranean the complex interactions between two

continental masses, the Eurasian and African plates, has produced a broad arcuate

collision zone, the Gibraltar Arc System (Fig. 1), comprised of the Betic and Rif

mountain ranges separated by the Alboran Sea basin. A wide variety of tectonic models

have been proposed, to explain the surface geology [see Platt et al., 2013 for a review]

whereas only recently has detailed information been available of the deep structure of

much of this region, including the Rif Cordillera in North Morocco.

Crustal thicknesses, composition, structure, and the location of major fault zones

reflect deformation processes, which is fundamental knowledge necessary to constrain

evolutionary models. Over 100 years after its discovery the depth of the Mohorovicic

discontinuity is still poorly known in many geodynamically complex areas [see

Carbonell et al., 2013 for a review]. The Gibraltar Arc System is one such area. Diaz

and Gallart [2009] and Gallart and Diaz [2013] reported the crustal thickness

measurements presently available in the Iberian Peninsula, revealing that the northern

3

part of this complex collision zone features crustal thicknesses between 30 and 43 km,

the latter found beneath specific areas of the Betic ranges. However, recent receiver

function analyses suggest greater crustal thicknesses in parts of the Betics and in the

Moroccan Rif. Here we present an analysis of wide-angle seismic data which allows the

development of a velocity-depth model of the crust and hence provide new insights on

the Moho topography beneath the area.

Numerous, often incompatible geodynamic models have been proposed to explain

the singular configuration of the Gibraltar Arc System. They include: regional-scale

recycling of the lithosphere by delamination [Seber et al., 1996]; slab break-off [Blanco

and Spakman, 1993; Zeck, 1996], convective removal [Platt and Vissers, 1989]; active

eastward subduction of oceanic crust [Gutscher et al., 2002]. Further details on those

models, including a summary diagram, can be found at the review paper presented by

Platt et al. [2013]. Many authors now relate the origin of the Gibraltar Arc to the

segmentation of the Western Mediterranean Subduction Zone (WMSZ) and the fast

westward oriented retreat of the subsequent narrow slab of oceanic lithosphere [Royden,

1993; Lonergan and White, 1997; Rosenbaum and Lister, 2004; Faccenna et al., 2004;

Jolivet et al., 2009; Vergs and Fernndez, 2012]. Recent mantle studies of this region

suggest subduction related convective recycling and delamination of mantle lithosphere

from the crust [Bezada et al, 2013; Palomeras et al, 2014; Thurner et al, 2014]. The

crustal models presented here helped constrain these recent studies by providing a

detailed knowledge of the crustal properties.

With an aim to improving the constraint on geodynamic models of the Gibraltar

Arc, an ambitious multidisciplinary research project was initiated in 2006, led by the

4

Spanish Topo-Iberia program in collaboration with PICASSO (Program to Investigate

Convective Alboran Sea System Overturn), a loosely affiliated consortium of U.S.,

Irish, German and Moroccan institutions. Within these initiatives special emphasis was

placed on geophysical characterization of the crust with active seismic and

magnetotelluric methods in specific areas coinciding with surface structural and

petrological studies. These extended from the center of the Iberian Peninsula [Gmez-

Ortiz et al., 2011; Pous et al., 2011; Martnez-Poyatos et al., 2012; Ruiz-Costn et al.,

2012; Ehsan et al., 2014; Garca-Lobn et al., 2014] to the Saharan craton, south of the

Atlas Mountains. A unique high resolution wide-angle reflection dataset was acquired

across the Atlas [Ayarza et al., 2014] and across the Rif [this manuscript].

We present seismic crustal velocity models along a 430 km-long and a 330 km-long

wide-angle seismic reflection (WA) transects crossing the Rif in the NS and EW

directions. This geometry is aimed to acquire seismic data over the zone where a

significant low Bouguer anomaly has been previously identified Hildenbrand et al.

[1988]. The Bouguer gravity anomaly database maintained by the International

Gravimetric Bureau (BGI; http://bgi.omp.obs-mip.fr/data-products/Grids-and-

models/wgm2012) shows a very prominent -150 mGal gravity low is located to the

south of the Internal Zones of the Rif, over the External Zones and the western region of

the Gharb foredeep basin (Fig. 2). The gravity anomaly increases towards the oceanic

areas, reaching maximum values of up to 250 mGal over the Atlantic and from 50 to

150 mGal in the Alboran Basin and its transition towards the oceanic Algerian basin

(Figs. 1 and 2). The new seismic models derived from our data are converted to density

and the predicted Bouguer anomaly is compared with the BGI database.

5

2. Geological Setting

The Gibraltar Arc system forms the Westernmost Mediterranean Alpine belt and

comprises the Betic and Rif Cordilleras and a deep sedimentary basin over the extended

continental crust of the Alboran Sea, which developed roughly synchronously with the

orogenic belt during the Miocene [Vergs and Fernndez, 2012; Platt et al., 2013].

Similar to the Betic Ranges and other Alpine Mediterranean cordilleras, the Rif

consists of Internal and External Zones, separated by Flysch units. The Internal zones

are formed by Paleozoic, Mesozoic and Cenozoic sequences, including metamorphic

complexes that have been affected by Alpine deformation since the Eocene-Late

Oligocene [Chalouan et al., 2001, 2008]. The External Zones comprise carbonate and

pelitic Mesozoic and Cenozoic units, mainly limestone and marls. In the Rif, they form

a fold-and-thrust belt detached along Late Triassic evaporite beds above the thinned

continental crust of the North Africa passive margin [Wildi, 1983; Chalouan et al.,

2008]. The Flysch units are composed of Cretaceous-Lower Miocene detrital rocks.

They overthrust the External Rif units and include klippes located on the Internal Zones

[Chalouan et al., 1995, 2008]. The relatively large Ronda and Beni-Bousera peridotite

exposures (Fig. 1) are found along the eastern flank of the Gibraltar Arc in the Betics

and the Rif, respectively.

The Gharb (or Rharb) Basin is a foredeep separating the Rif belt from the Moroccan

Meseta and Middle Atlas. This basin contains part of the Prerif nappe underlying a large

amount of continental sediments, reaching a maximum depth of 8 km towards the west

[Hafid et al., 2008]. It was moreover filled with sediments of marine origin during the

6

Tertiary and continental formations during the Quaternary, except for a coastal fringe

[Hafid et al., 2008]. The Gharb Basin, like its counterpart in the Iberian Peninsula, the

Guadalquivir Basin, evolved as a foreland basin as the basement was loaded by the

thrust sheets of the External Units [Fernndez et al., 1998; Garca-Castellanos, 2002]

during the Miocene.

In Early-Middle Miocene, after crustal thickening and metamorphism, the region

began to undergo EW to NE-SW extension that thinned the continental crust along

normal faults forming the Alboran Basin [Chalouan et al., 2008]. The Alboran Basin

has thick Neogene overlying deep crustal rocks locally recognized as belonging to the

Sebtides-Alpujarride nappes.

Since the Late Miocene, continued compression has formed large folds and reverse

faults in the mountain front, as well as normal faults in the upper crustal levels of the

Internal Zones, providing the relief of the modern Rif. Large strike-slip faults, notably

the Trans-Alboran Shear Zone (TASZ), and its onshore continuation, the Nekkor fault,

accommodate escape of Central Rif toward the SW [Prouse et al, 2010]. The TASZ is

a broad fault zone, composed of different left-lateral strike-slip fault segments running

from the eastern Betics to the Alhoceima region in the Rif and resulting in a major

bathymetric high in the Alboran Sea, that affects the Neogen basins of the region [Udas

and Buforn, 1992; Martnez-Daz et al., 2001]. The Nekkor fault is linked to the normal

faults beneath Alhoceimas region [Booth-Rea et al. 2012], which is one of the most

seismic activities areas nowadays in Morocco. More detailed descriptions of the

geology of the Gibraltar Arc can be found elsewhere [e.g. Chalouan et al., 2008; Platt et

al., 2013, and references therein].

7

3. Previous geophysical studies

Early geophysical studies of the Rif in the 70s consisting of low-density seismic

refraction by Hatzfeld and Bensari, [1977] reported a crustal thickness of 30 km beneath

the Gharb Basin. Beneath the Alboran Sea, wide-angle profiles (also in the 70's)

constrain Moho-depths of 18-20 km beneath the central part of the basin [Working

Group for Deep Seismic Sounding, 1978]. Wigger et al [1992] report crustal thicknesses

of 35 km beneath the southernmost Rif determined by refraction recordings of quarry

blasts.

Torne et al. [2000] used gravity, heat flow and elevation data to estimate crust and

mantle lithosphere thickness beneath the Alboran basin, the Betics and the Rif

Cordilleras. They found that crustal thickness between the orogenic belts and the

Alboran Sea may differ as much as 25 km. These models where further refined by

including constraints from the Geoid anomalies [Frizon de Lamotte et al., 2004; Zeyen

et al., 2005] and possible petrological compositions [Fullea et al., 2010, 2014]. These

resulted in a moderately thick crust underneath the Rif and Betics (~32-34 km), and a

thin continental crust (~18-22 km) beneath the Alboran basin. This basin progressively

thins towards the east, reaching values less than 16 km depth at the transition to the

Algerian basin. In the Moroccan interior, crustal thickness increases to 38 km below the

highest elevations of the High Atlas, then it decreases to the southeast to 30-32 km.

Results from a NW-SE oriented magnetotelluric profile across the Rif [Anahnah et

al., 2011a] show a heterogeneous upper crust, with resistive (metamorphic rocks) and

8

conductive (peridotites) bodies in the uppermost 10 km of the Internal Zones, and

highly conductive bodies in the External Zones and foreland basin. The variable

thicknesses of the latter ones suggest the presence of basement highs that may be related

to blind frontal thrusts between the Gharb Basin and the External Zones. A crustal

detachment level separating shallow geological units from a probable Variscan

basement was inferred. At depths below 5 km, relatively large resistive bodies appear

below the frontal part of the Rif. The southern Rif has also been modeled using

magnetotelluric data, featuring wide and thin conductive bodies interpreted to

correspond to detrital rocks that alternate with marl and carbonates [Anahnah et al.,

2011b].

Eurasian convergence relative to Africa trended south during the Late Cretaceous-

Paleogene, and has trended southeast oblique to the African margin from the Miocene.

The present-day tectonic motions in the study area are constrained from GPS

observations on permanent sites and temporary deployments [Fadil et al., 2006; Vernant

et al., 2010; Koulali et al., 2011] combined with other stress indicators [Palano et al.,

2013]. They show south to southeast motion at ~5 mm/year of the Rif region relative to

stable Nubia. The Rif, Betics and Alboran Sea are an active seismic zone, with the Rif

interpreted as a wide transpressive zone between the seismically active Tell to the east,

and the oceanic transform fault plate boundary to the west. This is interpreted as

additional evidence of subduction roll-back, corresponding to the contact area between

two converging plates, Eurasia and Africa.

Local earthquake travel time tomography using data from the TopoIberia and

PICASSO arrays of the Gibraltar Arc [El Moudnib et al., submitted] have significantly

9

enhanced the previous models presented by Gurra and Mezcua [2000] and Calvert et al.

[2000a]. At the uppermost crustal levels low velocities (5.5-5.75 km/s) are observed

beneath the Betics and the Alhuceimas region, while the velocities beneath the Rif

remain close to the 1D IGN reference model [Gurra and Mezcua, 2000]. At ~30 km

depth, a low velocity zone (6.3-6.7 km/s) clearly underlies the Gibraltar Arc from the

easternmost Betics to the southeastern Rif, were an abrupt change in velocity is

observed. Low shear velocities have been determined in the same region from Rayleigh

wave tomography [Palomeras et al, 2014]. Calvert et al. [2000b] and Serrano et al.

[2005] used seismic regional waveforms to map Pn velocities along the Africa-Iberia

boundary. A robust low-Pn (7.8 kms-1) velocity anomaly is imaged beneath the Betics,

in contrast with the relatively normal values beneath the Alboran Sea. The Rif and

Middle Atlas show also low Pn velocities. Diaz et al. [2013] resolved similar patterns

and a local area of high Pn and Sn velocity (8.2 kms-1 and 4.8 kms-1, respectively)

beneath the Alhoceimas region (approx. 35N, 3W).

Regional 3D teleseismic tomography [Bezada et al., 2013] images a high velocity

slab-like feature beneath the Alboran Sea and much of the eastern Betics from

lithospheric depths to >600 km. The geometry confirms slab tearing beneath the eastern

Betics, suggested previously by Spakman and Wortel [2004] and Garcia-Castellanos

and Villaseor [2011]. Rayleigh wave tomography and receiver function images of the

Western Mediterranean [Palomeras et al., 2014; Thurner et al, 2014] suggest that the

slab is attached to the crust beneath the Rif Cordillera, suggested also by Prouse et al.

[2010] from GPS observations.

SKS split directions rotate around the Gibraltar Arc with the fast directions

10

following the curvature of the Rif-Betic chain [Buontempo et al., 2008; Diaz et al.,

2010; Miller et al., 2013; Diaz and Gallart, 2014].This is interpreted as evidence of

asthenospheric mantle flow around the Alboran slab [Diaz et al., 2010; Alpert et al.,

2013; Diaz and Gallart, 2014]. Recent Topo-Iberia and PICASSO receiver function

studies [Mancilla et al., 2012; Thurner et al., 2014] show large variations in crustal

thickness beneath northern Morocco, with a clearly defined localized crustal root

beneath the central Rif extending to 40-50 km, and significantly thinner crustal

thicknesses of 22-30 km beneath northeastern Morocco, although the studies differ in

detail. The eastern limit of the Rif Cordillera, in the transition between both areas,

shows complex converted Ps signals admitting different interpretations. Thurner et al.

[2014] identified a subcrustal horizon beneath the Betic and Rif Cordilleras, located

between 45 and 80 km depth, which are interpreted as the top of the Alboran Sea slab

merging with the Moho at 50-55 km depth.

4. RIFSIS Data Acquisition and Processing

In October 2011 we acquired a 330 km-long and a 430 km-long wide-angle seismic

reflection profiles oriented, approximately, EW and NS (Fig. 1). The profile directions

were designed to approximate the overall Rif strike and dip directions and conform to

major and minor axes of the elliptical Bouguer anomaly pattern (Fig. 2). The EW

transect extends across the Rif orogen from the Gharb Basin to the Algerian border. The

NS line extended 70 km into the Iberian Peninsula and over 300 km within Morocco to

the Mid-Atlas, overlapping with the SIMA seismic wide-angle transect in the Atlas

Mountains [Ayarza et al., 2014]. Jointly, SIMA and the NS RIFSIS profile extend 700

km-long from the northern Sahara desert into southernmost Iberia. The Iberian portion

11

of the NS RIFSIS profile is not reversed as there were no source points in Iberia.

Each of the 5 sources consisted of 1Tn of chemical explosives in 2 boreholes and

was recorded by 845 digital seismographs with one-component 4.5 Hz geophones

(Reftek RF125 IRIS-PASSCAL Texans). The average receiver spacing was 750 m.

Shots R1 through R3 where located along the NS line, and R3-R5 were along the EW

line. Shot R3 is at the intersection of the two profiles. All shots were recorded by all the

stations producing fan shots for 3D control on deep structure [Carbonell et al., 2014].

Up to 402 seismographs were deployed along the EW profile and 443 along the NS

profile including 35 in Spain. The signal-to-noise ratio in our data is within the usual

range for this kind of experiments, providing a reasonably good overall data quality

although shot R1 near Gibraltar has low signal-to-noise ratio at offsets larger than 80

km.

Data processing included amplitude recovery, frequency filtering using a classical

band-pass Butterworth filter (3-10 Hz) and phase enhancement by a lateral phase

coherency filter [Schimmel and Gallart, 2007]. This latter procedure allows a better

identification of weak seismic arrivals at large offsets, as can be observed in the

Supplementary Figure S1. Travel times of different seismic phases were picked for a

series of crustal phases including the Moho reflection PmP. The coherency filter was

especially valuable in the case of shot R1 allowing us to identify arrivals from 80 km to

140 km offsets. A total of 2297 picks were obtained, 1154 from the NS profile and 1143

from the EW.

P-wave velocity-depth models were derived by forward modeling travel times s of

12

diving and reflected waves using RAYINVR software [Zelt and Smith, 1992].

Additional geological and geophysical constraints were considered in the modeling

procedure where available. For the NS profile we start from the velocity-depth model

presented by Ayarza et al. [2014] from the interpretation of the SIMA profile, crossing

the Atlas and partially overlapping our profile. At the northern edge, the seismic models

by Medialdea et al. [1986] were also taken into consideration. Different geological

results compiled in Chaloun et al. [2008] have been used to get an initial estimation of

the geometry and velocities in the uppermost sedimentary layers. Identified seismic

phases (shown and labeled in the figures) follow the conventional nomenclature: Ps and

Pg denote refractions through the sedimentary cover and the basement, respectively;

PiP, PcP and PmP stand for P to P reflections produced at the top of the middle crust,

top of the lower crust and Moho discontinuity, respectively; and P1, P1P identify

refraction and reflection events on a locally limited sedimentary layer. Because the shot

spacing is rather large, varying from about 80 km to 140 km, phases from the

sedimentary layers and the upper crust generally are not reversed, whereas the deeper

phases Pg, PcP, and PmP generally are. Despite careful inspection and analysis of the

filtered and unfiltered record sections, we have not found clear evidences of arrivals

displaying a convincing lateral correlation and which could be attributed to lower

crustal or Moho refracted phases. Hence, we have preferred to use only the well

identified reflected phases as the observations for modeling. In the following sections

we describe: 1) the quality of the data; 2) the parts of the models that are well

constrained and 3) the travel time picks and fit (Supplementary Figure S2). Estimated

uncertainties of the travel times picks are, on average 0.05 s for Ps, 0.1 s for Pg and

between 0.1-0.2 s for reflected phases PiP, PcP and PmP. The derived P-wave velocity

models reproduce the picked travel time branches with a very good agreement. The

13

observed travel time misfits (usually less than 0.15 s) are reasonable in view of

contributing factors such as the acquisition geometry, local oscillations on topography,

outcropping lithologies, etc.

5. Lateral variations of the structure beneath the External Rif domain: East-

West profile from Gharb Basin to the Algerian border

The EW profile is sampled from shot records R3, R4 and R5 (Figs. 3, 4 and 5).

Rapid structural variations beneath the External Rif domain can be inferred from the

deeper phases PcP and PmP recorded along this transect.

Shot R3 was recorded ~50 km to the west, towards the Gharb Basin, and 280 km to

the east, along the External Rif to the Algerian border. Two sedimentary arrivals are

observed in the eastern section: Ps1 is the first arrival to ~15 km and Ps2, to ~30 km

offset. A satisfactory fit in travel times is obtained (see Fig. 6) with a velocity gradient

between 3.2 and 3.8 km/s for the first sedimentary layer, and 4.25 to 5.2 km/s for the

second. In the western part, the first arrival to 15 km offset (-15 km in Fig. 3), identified

as the refracted phase P1, arrives earlier than the same arrival at similar offsets to the

east. This P1 phase, a first arrival to 40 km offset, travels in a layer with an average

velocity of 4.8 km/s. This phase and its associated reflection P1P indicate that the

Neogene deposits of the Gharb Basin extend to depths of 10 km. The Pg phase can be

correlated at offsets of 30 to 80 km in the eastern part with an apparent velocity of 5.8

km/s, but is less visible. PiP energy is observed beyond 30 km offsets at reduced times

of ~5 s to the east, almost a 1 s delay compared to the west. Weak PcP arrivals are

identified in the eastern section at offsets of 50-90 km, and relatively high amplitude

14

PmP arrivals at offsets of 80-190 km. When examined carefully, we identify two

arrivals with different slopes. The first event at 80-130 km offset arrives at ~10 s

reduced travel time. We identify a second event with velocity greater than 8.0 km/s at

110-140 km and 13-10 s travel times (Fig. 3). These two arrivals suggest a complex

Moho-topography. We have modeled the two arrivals as different PmP branches

resulting from a west to east shallowing of the Moho from more than 50 km to less than

30 km. The Moho ramp occurs over a distance of 45-50 km beneath the surface

expression of the Nekkor fault/TASZ, at the eastern end of the Bouguer gravity

anomaly over the Rif (Figs. 1, 2 and 6).

Shot R4, in the center of the EW profile (Figs. 1 and 4), clearly exhibits

extraordinary lateral differences in crustal structure beneath the Rif resulting from the

change in lower crustal structure. Sedimentary phases are observed as first arrivals to 20

km offsets to the west and 30 km to the east, suggesting that the sediments thin from

east to west from 4 to 2 km thickness. As will be discussed later, this step in the

sediment thickness is located beneath the surface expression of the Nekkor Fault. The

velocity in the sedimentary layers increases from 4.3 to 5.3 km/s. The Pg phase is

visible to 60 km offsets at apparent velocity 5.9 km/s. A reflected event (PiP),

appearing approximately symmetrically under the shotpoint from 20 to 100 km offset, is

interpreted to come from the top of the middle crustal layer at ~16 km depth. To

produce the relatively high amplitude of the PiP arrivals, a marked velocity contrast is

required, which we model as a transition from 5.9 to 6.3 km/s across the interface.

Hence, a 10 km thick upper crust is constrained under R4 in contrast with values of 5

km constrained beneath shotpoint R3 (Fig. 6).

15

The PcP phase is only observed east of R4 (Fig. 4), at around 5.5 s and at offsets of

60 km. Also to the east of R4, following the PcP phase we observe a high amplitude

PmP phase at 6.5 s (Fig. 4). The 1 s delay between both phases suggests the presence of

a rather thin, ~5 km lower crust to the east of R4. To the east, PmP is identified at

offsets greater than 70 km to the eastern end of the profile, modeled as a reflection from

a Moho at 30 km depth.

The most striking feature in the R4 shot record (Fig. 4) is the asymmetry observed

in PmP across the record section. PmP in the west appears from -170 to -60 km offset at

reduced travel times of 10-12 s, while in the eastern side it is observed around 7 s. For

an offset of 120 km, the difference between the PmP phase arrivals at both sides reaches

4 s (Fig. 4). This difference is accommodated in the final model by a 20 km difference

in crustal thickness (Figs. 4 and 6).

Shot record R5 (Fig. 5) reveals events from the sedimentary cover, middle crust,

lower crust and Moho. In detail a short sedimentary phase is observed to 10 km offset,

resulting from a thin (2-4 km) sedimentary layer with a velocity gradient of 3.0 to 3.8

km/s (Figs. 5 and 6). The Pg phase is correlated to 100 km offset with an average

velocity of 6.0 km/s. The PcP is visible after 50 km offsets (at about 5 s reduced time at

60 km) and the PmP after 60 km offsets, at about 2 s later on, constraining a Moho at

the eastern end of the profile at ~29 km depth (see Figs. 5 and 6). At offsets between

100-250 km, the observed arrivals from the deep crust are fitted in the model (see

Figure 5) as corresponding to PcP and refracted energy within the lower crust. They

could not been attributed to PmP to be consistent with the Moho location established

after the PmP fitting from reverse shot R4 (Figure 4). Hence, the lower crust in the east

16

is 7-8 km thick with a velocity gradient from 6.75 to 6.9 km/s, in strong contrast with

the western part of the profile, where the lower crust thickness exceeds 15 km, although

its top is poorly resolved from PcP phases.

Pn phases that would constrain the upper mantle velocities could not be identified

for any of the shots. This is not unusual in areas with significant lateral variations in the

crustal thickness, a weak velocity gradient at the uppermost mantle or a smooth crust-

mantle transition. However, this latter explanation seems not well in agreement with the

short PmP critical distances generally observed in our sections. We have fixed an

average velocity of 8 km/s in the uppermost mantle and verified that even if this value is

not constrained by direct Pn observations, a 2-3% modification of the velocity ratio

between the lower crust and mantle velocities results in clear misfits between observed

and theoretical PmP critical distances (Supplementary Figure S2).

In summary, in the crustal model along the EW RIFSIS transect (Fig. 6) the

thickness of sediments is clearly decreasing from west (Gharb Basin) to east (Algerian

border). Near R3 up to three sedimentary layers are interpreted in contrast to only one in

R5. Upper, middle and lower crustal levels are constrained by refracted and reflected

phases (Pg, PiP, PcP and PmP, respectively). The bottom of the crust is well defined

from PmP phases, although the lack of Pn arrivals prevents to constrain upper mantle

velocities. A major change in crustal thicknesses can be observed directly in the R4 shot

section, which shows differences of 5 s in the PmP arrivals at offsets of 120 km.

Forward modeling has shown that a rapid change of ~20 km in Moho depth within 50

km horizontal distances is needed to explain these observations. The Moho in this area

is sampled in both directions (shots R3 and R4) and hence the final model is here well

17

constrained. Maximum depths around 50 km are found beneath the high topography of

the External Rif domain, whereas a thin crust with a 29 km Moho depth is interpreted in

the foreland and Atlasic terranes up to the Algerian border.

6. North-South Rif Transect, from Middle Atlas domains to the Betics range

The North-South profile crosses the Internal and External Rif domains and the

transition to the Middle Atlas in the south and to the Betics in the north. This profile,

extending 430 km consisted of three shots, R1-R3 (Fig.1). Shotpoint R3 is at the

intersection of the NS and EW transects. Shot R1 was located 6 km south of the

Moroccan Mediterranean coast (Fig. 1). In Morocco all three shot records show two

sedimentary first arrivals: Ps1 to 10-20 km offset and Ps2 to 30-40 km (Figures 7-9).

North of the Gibraltar strait, the first station on the Iberian Peninsula is located at 35.5

km offset, and no sedimentary waves are visible on R1. Near R1 average velocities for

Ps1 and Ps2 are 4.0 to 4.8 km/s with layer thicknesses of 2 and 5 km, respectively (Figs.

7b and 10). The Pg phase is visible up to offsets of 60 km north and south, with an

average velocity of 5.8 km/s and intercept times of 4.5 to 5 s, probably due to the

relatively thick flysch terranes around the shotpoint. The PiP phase can be identified

from 30 to 80 km in both directions. Low amplitude arrivals characterize the PcP phase,

identified between 70-130 km to the south and between 60-80 km to the north. For both

PiP and PcP phases, arrival times are delayed ~1 s to the north relative to the south one,

suggesting 5-10 km thickening of the upper-middle crust beneath the Betics relative to

the northern Rif (Fig. 10). After applying lateral coherency filtering [Schimmel and

Gallart, 2007], PmP can be identified at offsets between 50-140 km to the south, and

between 60-90 km to the north. Arrival times are approximately the same for the same

18

offset, indicating a rather constant Moho depth, ~42 km, under the area sampled by R1

(Fig. 10).

The R2 shotpoint is located ~85 km to the south of R1 (Fig.1). The sedimentary

phases, Ps1 and Ps2, have velocity values and gradients similar to those at R1. The Pg

phase is identified to offsets of 80 km with an intercept time of 4.5 s. Intracrustal

reflections PiP and PcP are similar to R1 (Figs. 7 and 8). The PcP phase is rather weak

to the south, in contrast to higher amplitudes to the north to 140 km offset. PmP has the

highest amplitudes and is identified from 80-180 km offset to the south, at reduced

times of 11-8 s. PmP is not identified to the north.

R3 is located ~155 km from the R1 (Fig. 1). As in the other shot records, Ps1 phases

are first arrivals to offsets less than 18 km and are followed by a Ps2 phase which shows

clear differences in apparent velocities north and south (Fig. 9). The Ps2 phase to the

north has a higher apparent velocity than the corresponding phase to the south, and to

Ps2 on the other two shots R1 and R2. Shot R3 in the corresponding EW record section

indicated a relatively shallow (1.8 km) sedimentary interface separating Ps1 and Ps2,

giving rise to P1 and P1P in Fig. 3. Although these phases are not easily recognized in

the NS R3 record section (Fig. 9), the marked difference observed in the Ps2 apparent

velocities to the north and south is evidence for the locally limited sedimentary layer

from the EW profile.

Pg is clear and can be identified to 90 km offset south, and 60 km offset north (Fig.

9). The average apparent velocity is 5.7 km/s with rather large intercept times of 4.5 s

north and 5 s south. This constrains the thick Neogene basin below the shotpoint to ~5

19

km depth. The PiP phase appears from 40-100 km south, and 30-100 km north, with

similar arrival times indicating that the upper crust has a near constant thickness on both

side of shotpoint R3 (Fig. 9). In the model (Fig. 10) the basement is located at 5 km

depth, while the bottom of the upper crust is reached at 15 km. The PcP phase has been

interpreted only to the south (see Fig. 9). This phase is observed at offsets of 80-140 km

with low amplitude arrivals at 7.5 to 8 s respectively. The base of the crust is

constrained by the PmP phase, a relatively high amplitude arrival identified at times

around 8 s at offsets greater than 70 km south of R3, and with lower amplitudes from

100-160 km to the north, at reduced time of ~9.5 s. The time difference in PmP arrivals

is evidence of an increase in crustal thickness below R3, with the Moho at ~47 km

depth (Fig. 10). This is consistent with the R3 observations along the EW profile (Fig.

6). Additional constraints on the southern part of the NS profile are provided by the

SIMA project data, a similar wide-angle transect across the High and Middle Atlas that

overlapped the NS RIFSIS transect [Ayarza et al., 2014]. At the northern end of the

profile, in the Betics domain, constraint on the sedimentary structure and upper crust

were provided by Medialdea et al. [1986].

In summary, the crustal model along the NS RIFSIS transect consists of two

sedimentary layers with average velocities of 3.5 and 4.2 km/s inferred from the

observed first arrivals (Ps1, Ps2 and Pg). The upper, middle and lower crusts are

constrained by the refracted and reflected phases (Pg, PiP, PcP and PmP, respectively).

The average velocities within the crust are 5.9, 6.2 and 6.8 km/s. As for the East-West

profile we have assumed an 8 km/s velocity below the Moho. Crustal thickness reaches

43 km beneath the Betics and the Internal Rif domain, while under the External Rif

sampled area it increases to 47-49 km, with progressive thinning southward into the

20

Middle Atlas where the Moho is found at 31-32 km depth, consistently with the results

derived further South by Ayarza et al. [2014].

7. Gravity model

The marked lateral variations in crustal thickness inferred from the seismic models

are an outstanding structural feature that can be checked against the corresponding

gravity signature. The gravity anomaly beneath the Rif is remarkably low ~ -150mGal

(Fig. 2). These Bouguer values are of similar magnitude as those over the Betics and the

Atlas, where gravity lows are found over significantly higher elevations. The Rif gravity

low is an ellipse slightly shifted to the southwest of the Internal Zones, centered over

the External Zones and the western region of the Gharb foredeep basin (Figs. 1 and 2).

The latter is most likely related to the presence of a large volume (up to 8 km) of

continental sediments in the western part of the basin [Hafid et al., 2008].

To account for lateral variations perpendicular to the strike of the profiles, gravity

data have been averaged over a strip 25 km of the profiles with the resulting standard

deviation computed and displayed in Figures 11 and 12. The gravity models have been

built using the geometry derived from seismic modeling. An averaged P-wave velocity

value has been calculated for each layer in the seismic model and the corresponding

densities were calculated using the empirical relations by Brocher [2005], which applies

to most lithologies, except mafic- and calcium rich rocks. The crust is modeled as:

sedimentary layers with densities that range from 1800 to 2400 kg/m3, an underlying

crystalline basement with density of 2650 kg/m3, 2700 kg/m3 for the upper crust, 2800

to 2850 kg/m3 for the middle crust, 2850 to 2950 kg/m3 for the lower crust and 3300

21

kg/m3 for the uppermost lithospheric mantle. Calculations of the gravity model response

are based on the methods of Talwani et al. [1959], and Talwani and Heirtzler [1964],

and the algorithms described in Won and Bevis [1987]. The modeled crustal density

distribution is well compatible with the P-wave seismic velocity models. To achieve a

satisfactory fit to the gravity, the densities in the sedimentary cover needed to be

slightly adjusted locally.

For both transects (Figs. 11 and 12) the calculated gravity values lie very close to

average measurements and well within the standard deviation of the projected gravity

estimates. Particular attention has been taken in the segments directly sampled by

seismic reflections (thick black lines in Figs. 6 and 10) where we have kept the exact

geometry as obtained from the velocity model. In all cases the preferred option has been

to slightly change the density profile of the crustal layers instead of modifying their

geometry. Accordingly, the NS profile (Fig. 12) displays lateral density variations at

crustal levels from north to south. Densities slightly lower than expected are inferred in

the northern segment (from 0 to 140 km), in particular in the lower crust.

8. Discussion

The seismic refraction/wide-angle reflection experiment presented in this P-wave

study provides the previously poorly known first order crustal structure of the Rif

Mountains in northern Morocco. Crustal interface structure, particularly that of the

sedimentary basins and the Moho (Figs. 6 and 10) need to be taken into consideration in

any geodynamic model attempting to address the complex tectonic evolution of the

Gibraltar Arc System.

22

The western end of the EW profile, sampling the Gharb Basin, shows a thick

sedimentary cover reaching 10 km. At the transition between the External Rif domain

and the Gharb Basin the observation of a fast sedimentary layer westward of shot R3,

altogether with the significant change in the total sedimentary thickness, is interpreted

as being the deep expression of fold-and-thrust belt, which will reach at least depths

around 10 km. This is consistent with previous geological results showing the presence

of those structures at an analogue area between the Sass Basin and the External Rif

[Bargach et al., 2004; Chalouan et al., 2006]. Further East, the sedimentary cover thins

in a progressive manner to about 2 km near the point at which the Moho is deepest at

~53 km, increases abruptly to 4 km beneath shot R4 and remain 1-3 km thick up to the

Algerian border.

From the western end of the profile, the Moho deepens rather smoothly eastward to

model coordinate 145 km, where it thins abruptly from 53 km to less than 30 km by 190

km. That is, it decreases around 23 km in thickness over a distance of 40 km. This large

and unexpected variation in crustal thickness takes place under the External Rif domain.

Farther east, the crustal thickness between model points 190-330 km remains constant,

at ~29 km.

The abrupt change in crustal thickness between 145 and 190 km suggests a tectonic

boundary separating two different crustal types, which we attribute to juxtaposition of

crustal blocks along the seismically active left-lateral Nekkor strike-slip fault, the

onshore continuation of the TASZ. The low shot density and the obliqueness of the

profile across the shear zone limits our ability to resolve vertical features. However, a 2

23

km step in the sedimentary thickness, passing from 4 km eastward of the fault to 2 km

westward, has been documented. The Nekkor fault has been active from the late

Miocene [Chalouan et al., 2006] and has been linked to the normal faults beneath the

Alhoceimas region [Booth-Rea et al., 2012]. This region has large seismic activity in

the upper crust, but also moderate seismicity at depths reaching 40 km depth [El

Moudnib et al., submitted], hence suggesting that the Nekkor fault penetrates through

the entire crust. The abrupt change in the Moho depth beneath this area depicted by our

models strongly supports this hypothesis.

The density model along the EW transect is compatible with the gravity data.

Relatively high velocities and densities characterize the root zone. The high PmP

amplitudes under the root suggest that the Moho is a relatively sharp impedance

contrast, as smooth transition will result in changes in the critical distance inconsistent

with the data. The lower crust velocities in the southernmost part of the profile are

around 6.6 km/s, while beneath the Rif they reach values of 6.8 km/s. This suggests an

increase in metamorphic grade of the rocks in the root beneath the Rif. At this point it is

difficult to assess the increase of this metamorphic degree, as it could be characteristic

of the lower crustal rocks of the Rif crustal domain or it could result from the increased

pressures in the crustal root.

Our velocity-depth model evidences a significant mismatch between surface

topography and Moho geometry, as the major topographic elevations reaching 1700

meters, are found around shotpoint R4, coinciding not with the crustal root but with the

steep thinning Moho ramp. This effect has previously been observed in active orogens

as the Carpathians or the Caucasus and can be explained by an elastic effect, even if

24

other hypothesis, as the presence of mantle mechanisms that contribute to sustain

present-day topography cannot be discarded.

Along the NS crustal model the Moho shows also relatively large variations in

depth. In the Middle Atlas the Moho is constrained from SIMA wide-angle seismic

reflection data [Ayarza et al., 2014] and reaches values a little over 40 km depth (Fig.

10). The low-density lower crust inferred from the seismic model in this area suggests a

mantle contribution to sustain topography. Northward, the Moho shallows to 32 km

beneath the northern Meseta and the southern part of the External Rif, then thickens

again from model coordinate 140 to 50 km depth at model point 210 km forming the

root beneath the northern External Zone of the Rif.

Hence, the most conspicuous structural feature, well constrained from both profiles,

is the presence of a 47-53 km thick crust below the External Rif domain, which

conforms well to the overall gravity pattern. Moreover, clear lateral variations are

observed, particularly a thinning of the crust by ~20 km east of the Nekkor fault. A

similar crustal pattern was suggested in recent receiver functions analyses from Topo-

Iberia and PICASSO passive seismic surveys [Mancilla et al., 2012; Thurner et al.,

2014], as well as in the velocity anomaly variations inferred from local earthquake

tomography and ambient noise datasets [El Moudnib et al., submitted]. GPS

observations show that the Rif block is moving S to SW relative to Nubia, with an

eastward termination roughly coincident with the Nekkor fault [Kuolali et al., 2011]. At

the lithospheric scale other datasets such as SKS splitting [Diaz and Gallart, 2014] and

surface wave and teleseismic body wave tomography [Bezada et al., 2013; Palomeras et

al., 2014] also show first order discontinuities below this area. Our results are not

25

consistent with the Moho depth values of 32-36 km previously estimated from heat flow

data [Soto et al., 2008] or from combining elevation, geoid, gravity and petrological

constrains [Fullea et al., 2010, 2014].

We emphasized that the Moho depths of ~50 km found below the External Rif

domain are significantly greater than those below the Middle or High Atlas and most of

the Betic Range [Ayarza et al., 2014; Thurner et al., 2014]. This increase in crustal

thicknesses is accommodated at middle and lower crust levels, while the upper crust has

a homogenous thickness of 12-15 km. Our profiles do not provide enough control to

discern between a middle or lower crustal thickening (see Figures 6 and 10) Another

highlight of these profiles is the existence of a fast sedimentary layer beneath shotpoint

R3, the fast layer of 4.8 km/s and double thickness of sediments, is likely due to a

thrusting, which can be associated to a thrust front and fold axes, similar to what is

described by Bargach et al. [2004] and Chalouan [2006] in the Prerif area.

The complex crustal structure of the Rif domains is a consequence of the Miocene

collision between the Iberian and Africa plates combined with the westward rollback of

the Neo-Tethys slab. The crustal thinning observed east of the Nekkor fault may be

associated with the Neo-Tethys passive margin, the result of Mesozoic rifting [Gomez

et al., 2000; Tesn, 2009]. The crustal thickening under the External Rif can be

attributed to slab pull from the downgoing Alboran slab under the Gibraltar Arc System

which is imaged in a variety of tomographic images [Spakman and Wortel, 2004;

Garcia-Castellanos and Villaseor, 2011, Bezada et al., 2013; Palomeras et al., 2014].

However, these interpretations remain open questions to be developed in further

investigations.

26

9. Conclusions

The RIFSIS experiment consists of two wide-angle seismic profiles crossing the Rif

Cordillera along NS and EW trends. The 430 km-long NS line extends from the Middle

Atlas, across the Rif and Straits of Gibraltar into the Betic Ranges in southern Spain.

The 330 km-long EW profile starts at the Gharb Basin and extends to the Morocco-

Algeria border. The crustal structure revealed from 2D forward modeling of the seismic

data of both profiles delineates a complex, laterally-varying crustal structure that

distinguishes the crust of the Rif from surrounding tectonic units.

The velocity-depth models obtained from this experiment image a significant

crustal root, with Moho depth reaching 47-53 km beneath the External Rif domain. The

crust thins ~20 km over a horizontal distance of 40 km across the Nekkor fault on the

EW profile, and over a horizontal distance of 70 km towards the Middle Atlas on the

NS profile. This crustal structure identifies a Rif crustal block previously inferred from

geodetic data [Prouse et al., 2010]. We interpret this structure as resulting from a

complex interaction of the Miocene to present continent-continent collision between

Iberia and Africa plates, coupled with slab rollback of the Neo-Tethys Alboran slab.

These results are consistent with the overall Bouguer anomaly pattern, the recent

results from receiver function analyses and the velocity anomaly variations inferred

from tomographic inversions of local earthquakes, teleseismic body and Rayleigh waves

and ambient noise datasets [Mancilla et al., 2012; Palomeras et al., 2014;Thurner et al.,

2014; El Moudnib et al., submitted]. They should prove useful as constraints for future

27

geodynamic modeling on the evolution of the Gibraltar Arc System.

Acknowledgments

This research was funded by Spanish RIFSIS project CGL2009-09727. AG

benefited from a PhD grant associated to this project. Support was provided by Spanish

projects CSD2006-00041 Topo-Iberia and CGL2008-03474 Topo-Med, as well as

the NSF Continental Dynamics Program Grant EAR0808939. We thank many

colleagues from the Institut de Cincies de la Terra Jaume Almera (CSIC, Barcelona),

Institut Scientifique-Universit Mohammed V-Agdal, Rabat, Morocco and Rice

University, Texas, US for their help in the field experiment, and Pnina Miller and

Mickel Johnson from IRIS-PASSCAL for their processing support, and Bureau

Gravimtrique International (BGI)/IAG International Gravity Field Service

http://bgi.obs-mip.fr for facilitating the gravity data in the study area. We acknowledge

also Puy Ayarza for providing data from SIMA-Atlas experiment and subsequent

discussions, Concepcin Ayala for her assistance offered in GM-SYS and Jaume

Vergs for sharing his knowledge in the Moroccan structural geology. And revision of

anonymous reviewers helped to improve the quality of the manuscript. We appreciate

the support and technology of ICTJA-CSIC, Barcelona and Rice University, Houston

where modeling of these profiles was developed. The data used in this research is

available upon request to J. Gallart.

References

Alpert, L.A., M.S. Miller, T.W. Becker and A. Allam (2013), Structure beneath the Alboran

from geodynamic flow models and seismic anisotropy, J. Geophys. Res., 118, pp. 4265-

4277, doi:10.1002/jgrb.50309.

Anahnah F., J. Galindo-Zaldvar, A. Chalouan, J. Poum, P. Ruano, A, Pedrera, A. Ruiz-

Constn, M, Alhmamou, M. Benmakhlouf, P. Ibarra and E. Asensio (2011a), Crustal

resistivity structure of the southwestern transect of the Rif Cordillera (Morocco), Geochem.,

Geophys., Geosystem., 12, Q12016, doi:1029/2011GC003783

Anahnah, F., J. Galindo-Zaldivar, Chalouan, A. Pedrera, J. Ruano, J. Pous, W. Heise, A. Ruiz-

Constan, M. Benmakhlouf, A.C. Lpez-Garrido, M.F.Ahmamou, C. Sanz de Galdeano, J.

Arzate, P. Ibarra, L. Gonzlez-Castillo, N. Bouregba, F. Corbo, and E. Asensio (2011b),Deep

28

resistivity cross section of the intraplate Atlas Mountains (NW Africa): New evidence of

anomalous mantle and related Quaternary volcanism, Tectonics, 30, TC5014,

doi:10.1029/2010TC002859.

Ayarza, P., R. Carbonell, A. Teixell, I. Palomeras, D. Mart, A. Kchikach, M. Hanafi, A.

Levander, J. Gallart, M.L. Arboleya, J. Alcalde, M. Charroud and M. Amrhar (2014), Crustal

thickness and velocity structure across the Moroccan Atlas from long offset wide-angle

reflection seismic data: the SIMA experiment, Geochem., Geophys., Geosystem.,

doi:10.1002/2013GC005164

Bargach, K., P. Ruano, A. Chabli, J. Galindo-Zaldvar, A. Chalouan, A. Jabaloy, M. Akil, M.

Ahmamou, C. Sanz de Galdeano, and M. Benmaklouf (2004), Recent Tectonic Deformations

and Stresses in the Frontal Part of the Rif Cordillera and the Sass Basin (Fes and Rabat

Regions, Morocco), Pure appl. Geophys., 161, pp. 521-540, doi:10.007/s00024-003-2461-6

Bezada, M.J., E.D. Humphreys, D.R. Toomey, M. Harnafi, J.D. Dvila, and J. Gallart (2013),

Evidence for slab rollback in westernmost Mediterranean from improved upper mantle

imaging, Earth Planet. Sci. Lett., 368, pp. 51-60, doi:10.1016/j,espl.2013.02.024

Blanco MI, Spakman W (1993), The P wave velocity structure of the mantle below the Iberian

Peninsula: evidence for subducted lithosphere below Spain. Tectonophysics., 221, pp. 1334,

doi:10.1016/0040-1951(93)90025-F

Booth-Rea, G., A. Jabaloy-Snchez, A. Azdimousa, L. Asebriy, M. VzquezVilchez and M.

Martnez-Martnez (2012), Upper-crustal extension during oblique collision: the Temsamane

extensional detachment (eastern Rif, Morocco). Terra Nova, 24, pp. 505-512,

doi:10.1111/j.1365-3121.2012.01089.x

Brocher, TA (2005), Empirical relations between elastic wavespeeds and density in the earth's

crust. Bull. Seism. Soc. Am., 95, 6, pp. 2081-2092, doi:10.1785/01200050077 Buontempo, L., G.H.R Bokelmann, G. Barruol and J. Morales (2008), Seismic anisotropy

beneath southern Iberia from SKS splitting, Earth Planet. Sci. Lett.,273, pp. 237-250,

doi:10.1016/j.epsl.2008.06.024

Calvert, A., E. Sandvol, D. Seber, M. Barazangi, S. Roecker, T. Mourabit, F. Vidal, G.

Alguacil, and N. Jabour (2000a), Geodynamic evolution of the lithosphere and upper mantle

beneath the Alboran region of the western Mediterranean: Constraints from travel times

tomography, J. Geophys. Res., 105, pp. 10,871-10,898.

Calvert, A., E. Sandvol, D. Seber, M. Barazangi, F. Vidal, G. Alguacil and N. Jabour (2000b),

Propagation of regional seismic phases [Lg and Sn] and Pn velocity structure along the

AfricaIberia plate boundary zone: tectonic implications. Geophys. J. Int. 142, pp. 384408

Carbonell, R., A.Levander and R. Kind (2013), The Mohorovicic discontinuity beneath the

continental crust: An overview of seismic constraints Tectonophysics., 609, pp. 353-376,

doi:10.1016/j.tecto.2013.08.037

29

Carbonell, R., P. Ayarza, J.Gallart, J. Diaz, M. Harnafi, A. Levander, and A.Teixell (2014),

From the Atlas to the Rif a Crustal seismic image across Morocco: The SIMA & RIFSEIS

control source wide-angle seismic reflection data. Paper presented at EGU General

Assembly, European Geosciences Union, Vienna, Austria

Chalouan, A., A. Ouazani-Touhami, L. Mouhir, R. Saji and M. Benmakhlouf (1995), Les failles

normales a faible pendage du Rif interne (Maroc) et leur effet sur l'amincissement crustal du

domaine D'Alboran, Geogaceta, 17, 3.

Chalouan, A., A. Michard, H. Feinberg, R. Montigny, and O. Saddiqi (2001), The Rif mountain

building (Morocco): a new tectonic scenario, Bull. Sec. geol. France, 172(5), pp. 603-616.

Chalouan, A., J. Galindo-Zaldvar, M. Akil, A. Chabli, P. Ruano, K. Bargach, C. Sand de

Galdeano, M. Benmakjlouf, M. Ahmamou and L. Gourari (2006), Tectonic wedge escape in

the southwestern front of the Rif Cordillera (Morocco), in Tectonics of the Western

Mediterranean and North Africa, edited by Moratti, G and A. Chalouan, pp. 101-118,

Geological Society, London

Chalouan, A., A. Michard, Kh. El Kadiri, F. Negro, D. Frizon de Lamotte, J.I. Soto, and O.

Saddiqi (2008), The Rif Belt, in Continental Evolution: The Geology of Morocco, edited by

Michard, A., O. Saddiqi, A. Chalouan, and D. Frizon de Lamotte, pp. 203-302, Springer,

Berlin

Diaz, J. and J. Gallart (2009), Crustal structure beneath the Iberian Peninsula and surrounding

waters: A new compilation of deep seismic sounding results. Phys. Earth Planet. Int. 173, pp.

181-190, doi:10.1016/j.pepi.2008.11.008

Diaz, J., J. Gallart, A. Villaseor, F. Mancilla, A. Pazos, D. Crdoba, J.A. Pulgar, P. Ibarra and

M. Harnafi (2010),Mantle dynamics beneath the Gibraltar Arc (western Mediterranean) from

shear-wave splitting measurements on a dense seismic array, Geophys. Res. Lett., 37,

L18304, doi:10.1029/2010GL044201

Diaz, J., A. Gil, and J. Gallart (2013), Uppermost mantle seismic velocity and anisotropy in the

Euro-Mediterranean region from Pn and Sn tomography, Geophys. J. Int., 192, pp. 310-325,

doi:10.1093/gji/ggs016

Diaz J., and J. Gallart (2014), Seismic anisotropy from the Variscan core of Iberia to the

Western African Craton: New constrains on upper mantle flow at regional scales. Earth

Planet. Sci. Lett., 394, pp. 48-47, doi:10.1016/j.epsl.2014.03.005

Ehsan, S.A., R. Carbonell, P. Ayarza, D. Mart, A. Prez-Estan, D.J. Martnez-Poyatos, J.F.

Simancas, A. Azor and L. Mansilla (2014), Crustal deformation styles along the reprocessed

deep seismic reflection transect of the Central Iberian Zone (Iberian Peninsula),

Tectonophysics, 621, pp. 159-174, doi:10.1016/j.tecto.2014.02.014

El Moudnib, L., A. Villasenor, M. Harnafi, M.M.Himmi and J. Gallart (submitted), Crust

structure of northern Morocco and southern Iberian Peninsula from local earthquake

30

tomography, Tectonophysics

Faccenna, C., C. Piromallo, A. Crespo-Blanc, L. Jolivet, F. Rossetti (2004), Lateral slab

deformation and the origin of the Western Mediterranean arcs, Tectonics 23,

doi:10.1029/2002TC001488.

Fadil, A., P. Vernant, S. McClusky, R. Reilinger, F. Gomez, D. B. Sari, T. Mourabit, K. L.

Feigl, and M. Barazangi (2006), Active tectonics of the western Mediterranean: GPS

evidence for roll back of a delaminated sub-continental lithospheric slab beneath the Rif

mountains, Geology, 34, pp. 529532, doi:10.1130/G22291.1.

Fernndez, M., X. Berstegui, C. Puig, D. Garcia-Castellanos, M.J. Jurado, M. Torne, and C.

Banks (1998) , Geophysical and geological constraints on the evolution of the Guadalquivir

foreland basin, Spain, in Cenozoic Foreland Basins of Western Europe, edited by Mascle, A.,

C. Puigdefabregas, H.P. Luterbacher, M. Fernndez, pp. 29-48 Geological Society, London

Frizon de Lamotte, D., A. Crespo-Blanc, B. Saint-Bzar, M. Comas, M. Fernndez, H. Zeyen,

H. Ayarza, C. Robert-Charrue, A. Chalouan, M. Zizi, A. Texell, M.L. Arboleya, F. Alvarez-

Lobato, M. Julivert, A. Michard, TRASNSMED-transect I [Betics, Alboran Sea, Rif,

Moroccan Meseta, High Atlas, JbelSaghro, Tindouf Basin] (2004),in The TRANSMED

Atlas - the Mediterranean Region from Crust to Mantle edited by Cavazza, W., F.M., Roure,

W., Spakman, G.M., Stampfli, and P.A., Ziegler, etal. pp. 91-96Springle, Berlin

Fullea, J., M. Fernndex, J.C. Afonso, J. Vergs, and H. Zeyen (2010), The structure and

evolution of the lithosphere-asthenosphere boundary beneath the Atlantic-Mediterranean

Transition Region, Lithos, 120 (1-2), pp. 74-95, doi:101016/j.lithos.2010.03.003

Fullea, J., J. Rodrguez-Gonzlez, M. Charco, Z. Martinec, A. Negredo, and A. Villaseor

(2014), Perturbing effects of sub-lithospheric mass anomalies in GOCE gravity gradient and

other gravity data modelling: Application to the Atlantic-Mediterranean transition zone. Int.

J. Appl. Earth Observ. Geoinf. doi:10.1016/j.jag.2014.02.003

Gallart, J. and J. Diaz (2013), Outstanding Moho depth variations in the Iberian Peninsula, NW

Africa and surrounding margins, revealed from controlled-source seismic surveys. Paper

presented at EGU General Assembly, European Geosciences Union, Vienna, Austria

Garcia-Castellanos, D. (2002), Interplay between lithospheric flexure and river transport

foralend basins. Basin Res. 14 (2), pp. 89-104, doi:10.1046/j.1365-2117.2002.00174.x

Garcia-Castellanos, D., and A. Villaseor (2011), Messinian salinity crisis regulated by

competing tectonics and erosion at the Gibraltar Arc. Nature 480, pp. 359-363,

doi:10.1038/nature10651

Garca-Lobn, C. Rey-Moral, C. Ayala, L.M. Martn-Parra, J. Matas and M.I. Reguera (2014),

Regional structure of the southern segment of Central Iberian Zone (Spanish Variscan Belt)

interpreted from potential field images and 2.5 D modelling of Alcudia gravity transect,

Tectonophysics, 614, pp. 185-202, doi:10.1016/j.tecto.2013.12.005

31

Gomez, F., W. Beauhamp, and M. Barazangi (2000), Role of the Atlas Mountains (northwest

Africa) within the African-Eurassian plate-boundary zone. Geology, 28(9), pp.775-778,

doi:10.1130/0091-7613

Gmez-Ortiz, D., B. N. P. Agarwal, R. Tejero and J. Ruiz (2011), Crustal structure from gravity

signatures in the Iberian Peninsula. GSA Bulletin vol. 123, no. 7/8, pp. 1247-1257,

doi:10.1130/B30224.1

Gurra E., and J. Mezcua (2000), Seismic tomography of the crust and lithospheric mantle in the

Betic Cordillera and Alboran Sea. Tectonophysics, 329, pp. 99119

Gutscher, M.A., J. Malod, J.P. Rehault, I. Contrcci, F. Klingelhoefer, L. Mendes-Victor, and W.

Spakman (2002), Evidence for active subduction beneath Gibraltar, Geology, 30, 1071-1074

Hafid, M., G. Tari, D. Bouhadioui, I. El Moussaid, H. Echarfaoui, A. Ait Salem, M. Nahim, and

M. Dakki (2008), Atlantic Basins, in Continental Evolution: The Geology of Morocco edited

by Michard, A., O. Siddiqi, A. Chalouan, and D. Frizon de Lamotte, pp. 303-330, Springer,

Berlin

Hatzfeld, D., and D. Bensari (1977), Grands profile sesmiques dans la rgion de l'arc de

Gibraltar, Bull. Sec. Gol. France, 7, XIX-4, pp. 749-756

Hildenbrand, T. G., R. P. Kucks, M. F. Hamouda, and A. Bellot (1988), Bouguer gravity map

and related filtered anomaly maps of Morocco, U.S. Geol. Surv. Open File Rep., pp. 88-517.

Jolivet, L., C. Faccenna and C. Piromallo (2009), From mantle to crust: Stretching the

Mediterranean. Earth Planet. Sci. Lett., 285, pp. 198209, doi:10.1016/j.epsl.2009.06.017

Koulali, A., D. Ouazar, A. Tahayt, R.W., King, P. Vernant, R.E. Reilinger, S. McClusky, T.

Mourabit, J.M. Davila, and N. Amraoui (2011), New GPS constraints on active deformation

along the Africa-Iberia plate boundary, Earth Planet. Sci. Lett., 308, pp. 211-217,

doi:10.1016/j.epsl.2011.05.048

Lonergan, L., and N. White (1997), Origin of the Betic-Rif mountain belt, Tectonics, 16, pp.

504-522, doi:10.1029/96TC03937

Mancilla, F.L., D. Stich, J. Morales, J. Juli, J, Diaz, A. Pazos, D. Crdoba, J.A. Pulgar, P.

Ibarra, M. Harnafi, and F. Gonzalez-Lodeiro (2012), Crustal thickness variations in Northern

Morocco, J. Geophys. Res., 177, B02312, doi:10.1029/2011JB008608

Martnez-Daz, JJ., E. Masana, JL. Hernndez-Enrile, P. Santasach (2001), Evidence for

coseismic events of recurrent prehistoric deformation along the Alhama de Murcia fault,

southern Spain, Acta Geologica Hipanica, 36 (3-4), 12

Martnez-Poyatos, D., R. Carbonell, I. Palomeras, J. F. Simancas, P. Ayarza, D. Mart, A. Azor,

A. Jabaloy, P. Gonzlez Cuadra, R. Tejero, L. M. Martn Parra, J. Matas, F. Gonzlez

Lodeiro, A. Prez-Estan, J. L. Garca Lobn and L. Mansilla (2012), Imaging the crustal

structure of the Central Iberian Zone (Variscan Belt): the ALCUDIA Deep seismic reflection

transect. Tectonics, 31, p. TC3017, doi:10.1029/2011TC002995

32

Medialdea, T., E. Suriach, R. Vegas, E. Banda, and J. Ansorge (1986), Crustal structure under

the western end of the Betic cordillera (Spain), Annales Geophysicae, 4 (B4), pp. 457-464.

Miller, M.S., A. Allam, T.W. Becker, J.F. Dileo, and J. Wookey (2013), Constraints on the

tectonic evolution of the westernmost Mediterranean and northwestern Africa from shear

wave splitting analysis. Earth Planet. Sci. Lett.. doi:10.1016/j.epsl.2013.05.036

Palano, M., P. Gonzalez and J. Fernandez (2013), Strain and stress fields along the Gibraltar

Orogenic Arc: constraints on active geodynamics Gondwana Res., 23 (2013), pp. 10711088

doi:10.1016/j.gr.2012.05.021

Palomeras, I., S. Thurner, A. Levander, K. Liu, A. Villaseor, R. Carbonell and M. Harnafi

(2014), Finite-frequence Rayleigh wave tomography of the western Mediterranean: Mapping

its lithospheric structure, Geochem., Geophys., Geosystem., 15 (1), pp. 140-160,

doi:10.1002/2013GC004861

Prouse, E., P. Vernant, J. Chery, R. Reilinger, and S. McClusku (2010), Active surface

deformation and sub-lithospheric processes in the western Mediterranean constrained by

numerical models, Geology 38(9), pp. 823-826, doi:10.1130/G30963.1

Platt, J.P. and R.L.M. Vissers (1989), Extensional collapse of thickened continental lithosphere:

An hypothesis for the Alboran Sea and Gibraltar arc, Geology, 17, pp. 540-543.

Platt, JP., W.M. Behr, K. Johanesen and J.R. Williams (2013), The Betic-Rif Arc and Its

Orogenic Hinterland: A Review. Annu. Rev. Earth Planet. Sci., 41, pp.313357,

doi:10.1146/annurev-earth-050212-123951

Pous, J., D. Martnez-Poyatos, W. Heise, F. Monteiro Santos, J. Galindo-Zaldvar, P. Ibarra, A.

Pedrera, A. Ruiz-Constn, F. Anahnah, R. Gonalves and A. Mateus (2011), Constraints on

the crustal structure of the internal Variscan Belt in SW Europe: A magnetotelluric transect

along the eastern part of Central Iberian Zone, Iberian Massif, J. Geophys. Res., 116,

B02103, doi:10.1029/2010JB007538.

Rosenbaum, G.andLister, G.S., (2004), Neogene and Quaternary rollback evolution of the

Tyrrhenian Sea, the Apennines, and the Sicilian Maghrebides.Tectonics 23, TC1013,

doi:10.1029/2003TC001518.

Royden, L.H. (1993), Evolution of retreating subduction boundaries formed during continental

collision, Tectonics, 12, pp. 629-638, doi:10.1029/92TC02641

Ruiz-Constn, A., A.Pedrera, J.Galindo-Zaldvar, J.Pous, J.Arzate, F.J.Roldn-Garca, C.Marin-

Lechado, and F.Anahnah (2012), Constraints on the frontal crustal structure of a continental

collision from an integrated geophysical research: The central-western Betic Cordillera (SW

Spain), Geochem.,Geophys., Geosyst., 13, Q08012, doi:10.1029/2012GC004153.

Seber, D., M. Barazangi, B. A. Ibenbrahim, and A. Demnati (1996), Geophysical evidence for

lithospheric delamination beneath the Alboran Sea and Rif-Betic Mountains, Nature, 379, pp.

785-790, doi:10.1038/379785a0

33

Serrano, I., T.M. Hearn, J. Morales and F. Torcal (2005), Seismic anisotropy and velocity

structure beneath the southern half of the Iberian Peninsula. Phys. Earth Planet.Int, , 150, pp.

317-330, doi:10.1016/j.pepi.2004.12.003

Schimmel, M., and J. Gallart (2007), Frequency-dependent phase coherencefor noise

suppression in seismic array data, J. Geophys, Res., 112, B04303, doi:1029/2006JB0046860

Soto, J.I., F. Fernndez-Ibez, M. Fernndez, and A. Garca-Casco (2008), Thermal structure

of the crust in the Gibraltar Arc: Influence on active tectonics in the western Mediterranean,

Geochem., Geophys., Geosystems, 9(10), pp. 1525-2027, doi:10.1029/2008GC002061

Spakman, W. and M.J.R. Wortel (2004), A tomographic view on western Mediterranean

geodynamics, in The TRANSMED Atlas The Mediterranean region from crust to mantle.

edited by Ziegler, P., pp. 31-52, Springer, Berlin

Talwani, M., Heirtzler, J. R. 1964. Computation of magnetic anomalies caused by two

dimensional bodies of arbitrary shape. In: Parks, G. A. (Ed.), Computers in the

mineral industries, Part 1. Stanford University Publications, Geological Sciences, 9:

464-480.

Talwani, M., Worzel, J. L., Landisman, M. 1959. Rapid gravity computations for two

dimensional bodies with application to the Mendocino submarine fracture zone.

Journal of Geophysical Research, 64: 49-59. Tesn, E. (2009), Estructura y cronologa de la deformacin en el borde Sur del Alto Atlas de

Marruecos a partir del registro tectono-sedimentario de la cuenca de antepas de Ouarzazate,

Ph.D. Thesis, UniversitatAutnoma de Barcelona, Spain

Thurner, S., I. Palomeras, A. Levander, R. Carbonell and L. Cin-ty (2014), Evidence for

Ongoing Lithospheric Removal in the Western Mediterranean: Ps Receiver Function Results

from the PICASSO Project, Geochem., Geophys., Geosystems, doi:10.002/201GC005124

Torne, M., M. Fernndez, M. C. Comas, and J. I. Soto (2000), Lithospheric structure beneath

the Alboran Basin: Results from 3D gravity modeling and tectonic relevance, J. Geophys.

Res., 105(B2), pp. 32093228, doi:10.1029/1999JB900281.

Udas, A and E. Buforn (1992), Sismicidad y sismotectnica de las Beticas, Fsica de la Tierra,

4, pp. 109-123

Vergs, J., and M. Fernndez (2012), Tethys-Atlantic interaction along the Iberia-Africa plate

boundary: The Betic-Rif orogenic system, Tectonophysics, 579, pp. 144-172,

doi:101016/j.tecto.2012.08.032

Vernant, P., A. Fadil, T. Mourabit, D. Ouazar, A. Koulali, J.M. Davila, J. Garate, S. McClusky,

and R. Reilinger (2010), Geodetic constraints on active tectonics of the Western

Mediterranean: Implications for the kinematics and dynamics of the Nubia-Eurasia plate

boundary zone, J. Geodynamics, 49, pp. 123-129, doi:10.1016/j.jog.2009.10.007

34

Wigger, P., G. Asch, P. Giese, W. D. Heinsohn, S. O. E. Alami, and F. Ramdami (1992),

Crustal structure along a traverse across the Middle and High Atlas mountains derived from

seismic refraction studies, Geol. Rundsch., 81(1), pp. 237248, doi:10.1007/BF01764552.

Wildi, W. (1983), La chanetello-rifaine (Algrie, Maroc, Tunisie): Structure, stratigraphie

etvolution du Trias au Miocne, Rev. Dyn. Geogr. Phys., 24, pp. 201-297

Won, I.J., Bevis, M. 1987. Computing the gravitational and magnetic anomalies due to a

polygon: Algorithms and Fortran subroutines. Geophysics, 52: 232-238.Working Group For

Deep Seismic Sounding In Alboran 1974 (1978), Crustal seismic profiles in the Alboran sea-

preliminary results. Pageoph, 116, pp. 166180.

Zeck, H.P. (1996),Betic-Rif orogeny: Subduction of Mesozoic Tethys lithosphere under

eastward drifting Iberia, slab detachment shortly before 22 Ma and subsequent uplift and

extensional tectonics, Tectonophysics, 254, pp. 1-16, doi:10.1016/0040-1951(95)00206-5

Zelt, C.A, and R.B. Smith (1992), Seismic traveltime inversion for 2-D crustal velocity

structure, Geophys. J. Int., 108, pp. 16-34, doi:10.1111/j.1365-246x.1992.tb00836.x

Zeyen, H., P. Ayarza, M. Fernndez, and A. Rimi (2005), Lithospheric structure under the

western African-European plate boundary: A transect across the Atlas Mountains and the

Gulf of Cadiz, Tectonics, 24, TC2001, doi:10.1029/2004TC001639

Figure Captions Figure 1.Map of Southern Iberia and Northern Morocco with the location of seismic wide-angle profiles acquired through the RIFSIS project (in red, the digital stations;

stars, the source points) and simplified geology of the study area. The major tectonic

domains and boundaries are indicated. GB: Gharb Basin; GuaB: Guadalquivir Basin;

GS: Gibraltar Strait; M: Meseta; MA: Middle Atlas; Nf: Nekkor fault; SB: Sass Basin;

T: Tell Mountain TASZ: Trans-Alboran Shear Zone. The dashed line represents the

geometry of the Gibraltar Arc The inset shows location within the Euro-Mediterranean

domain and includes an outline of the Westernmost Mediterranean Alpine Belt.

Figure 2.Gravity anomaly contour map of the southern Iberian Peninsula and northern Morocco. Bouguer anomalies have been extracted from the BGI (Bureau Gravimtrique

International, http://bgi.obs-mip.fr). The contour interval is 40 mGal. The red lines

indicate the location of the seismic profiles and the yellow stars represent the source

points. Note that the wide-angle transects closely follow the axes of the minimum of the

Bouguer anomaly (-150 mGal) beneath the Rif domain.

35

Figure 3. Shot R3 recorded along the EW profile with the ray tracing diagram and the travel time picks considered with topography and geological regions on top. a) R3 shot

record, applying a reduction velocity of 6 km/s and a bandpass filter of 3-10 Hz,

showing the quality of the data and the phases identified (see label description in text)

and used to build up the seismic velocity model. b) Ray tracing diagram revealing the

extension of the interfaces constrained in the final model. c) Fitting between travel time

picks determined from the data (in colors according to their travel paths) and the ones

modeled (in black) by the RAYINVR package [Zelt and Smith, 1992]. The vertical bars

associated to the travel time picks are indicative of the error estimates considered for

each pick. Colors meaning: red: Ps1 phase; green Ps2; light green: P1; dark green: P1P;

dark blue: Pg; light blue: PiP; purple: PcP; orange: PmP.

Figure 4. Shot R4 recorded along the EW profile with the ray tracing diagram and the travel time picks considered. See caption in Fig. 3 for details.

Figure 5. Shot R5 recorded along the EW profile with the ray tracing diagram and the travel time picks considered. See caption in Fig. 3 for details.

Figure 6.Crustal velocity-depth model determined along the EW profile with a vertical exaggeration of 2:1. The geologic domains are labeled on top of the model, and the

topographic profile is sketched. Bold lines indicate layer segments directly sampled by

seismic reflections. The vertical rectangles indicate the position where 1D velocity-

depth functions have been estimated, beneath the Gharb Basin (brown), the External Rif

domain (green) and the former North African margin (blue).

Figure 7. Shot R1 recorded along the NS profile with the ray tracing diagram and the travel time picks considered. See caption in Fig. 3 for details.

Figure 8. Shot R2 recorded along the NS profile with the ray tracing diagram and the travel time picks considered. See caption in Fig. 3 for details.

Figure 9. Shot R3 recorded along the NS profile with the ray tracing diagram and the travel time picks considered. See caption in Fig. 3 for details.

36

Figure 10.Crustal velocity-depth model determined along the NS profile, with a vertical exaggeration of 2:1. The geologic domains are labeled on top of the model, and the

topographic profile is sketched. Bold lines indicate layer segments directly sampled by

seismic reflections. The vertical rectangles indicate the position where 1D velocity-

depth functions have been estimated, beneath the Moroccan Meseta (brown), the

External Rif domain (green) and the Internal Rif domain (blue).

Figure 11.Crustal density model obtained by gravity forward modeling along the EW transect. a) 2D projection of the gravity measures extracted from the gravity anomaly

map (Fig. 2) along a 50 km wide strip. The vertical bars denote standard variation at

each measure point, and the red line denote the data fitting with the 2D gravity profile

generated by the density crustal model; b) distribution of the densities within the crust,

inferred from the seismic model (Fig. 6). Numbers are density values in (g/cm3). The

geometry for the crustal bodies has been kept and the density structure has been slightly

modified to improve the fitting.

Figure 12.Crustal density model obtained by gravity forward modeling along the NS transect. a) 2D projection of the gravity measures extracted from the gravity anomaly

map (Fig. 2) along a 50 km wide strip. The vertical bars denote standard variation at

each measure point, and the red line denote the data fitting with the 2D gravity profile

generated by the density crustal model; b) distribution of the densities within the crust,

inferred from the seismic model (Fig. 10). Numbers are density values in (g/cm3). The

geometry for the crustal bodies has been kept and the density structure has been slightly

modified to improve the fitting.

Internal Zones

Flysches

External Zones

Neogene Basins

Meso-Cenozoic plateaux

Sahara Domain

Atlas Mountains

Shot

Peridotites massif

Profiles

Figure 1

GS

Gulf of Cadiz

TASZ

Nf

Alboran Sea

RifGB

SB

Betic

M MA

T

GuaB

ALGERIA

MOROCCO

MEDIT

ERRA

NEAN

SEA

ATLANTIC OCEAN

Algeri

an B

asin

-8 -6 -4 -2

32

34

36

-120

-120

-120

-120

-80

-80

-80

-80

-80

-80

-80

-40

-40

-40

-40

-40

-40

0

0

0

0

0

40

40

40

80

80

120

-200 -160 -120 -80 -40 0 40 80 120 160 200 240 280

R1

R2

R3

R4

R5

Figure 2

Gravity anomaly (mGal)

01

23

45

67

89

10

111

21

31

41

5R

ed

uce

d tim

e [s

] Vr=

8 k

m/s

-40

-20 0

20

40

60

80

10

0

12

0

14

0

16

0

18

0

20

0

22

0

24

0

26

0

Distance [km]

RifS

eis

-R3

-EW

R3

Ps2

PmP

Pg

Ps1

PcP

Ps

P1

P1P

1

PiP

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

60

50

40

30

20

10

0

DE

PT

H (

km)

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

14

12

10

86

42

0

T-D

/8 (

s)

Figure 3

a

b

c

R3

W EGharb Basin

External Rif domain Former North African marginE

lev.

(km

) 1.51.0

0.5

0.0

01

23

45

67

89

10

111

21

31

41

5R

ed

uce

d tim

e [s

] Vr=

8 k

m/s

-18

0

-16

0

-14

0

-12

0

-10

0

-80

-60

-40

-20 0

20

40

60

80

10

0

12

0

14

0

Distance [km]

RifS

eis

-R4

-EW

R4

Pg

Pg

PcP

PiP

PmP

PmP

Ps1 Ps1Ps2

Ps2

PiP

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

60

50

40

30

20

10

0

DE

PT

H (

km)

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

14

12

10

86

42

0

T-D

/8 (

s)

a

b

c

Figure 4

R4

W EGharb Basin

External Rif domain Former North African marginE

lev.

(km

) 1.51.0

0.5

0.0

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

14

12

10

86

42

0

T-D

/8 (

s)

01

23

45

67

89

10

1112

13

14

15

Reduced tim

e [s

] Vr=

8 k

m/s

-320

-300

-280

-260

-240

-220

-200

-180

-160

-140

-120

-100

-80

-60

-40

-20 0

Distance [km]

RifS

eis

-R5

-EW

R5

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

60

50

40

30

20

10

0

DE

PT

H (

km)

PmP

PcP

PiP

Pg

Ps1

Figure 5

a

b

c

R5

W EGharb Basin

External Rif domain Former North African marginE

lev.

(km

) 1.51.0

0.5

0.0

60

40

20

0

2 4 6 8

Vp (km/s)

de

pth

km

-60

-50

-40

-30

-20

-10

0

de

pth

(km

)

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

Distance (km)

-2-4-6-8

32

34

36

W EGharb Basin

External Rif domainNekkor fault

Former North African margin

Ele

v. (

km)

1.5

1.0

0.5

0.0

Figure 6

R3 R4 R5

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5

velocity (km/s)

DISTANCE (km)

14

12

10

86

42

0

T-D

/8 (

s)

340 360 380 400 4200

20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

01

23

45

67

89

10

111

21

31

41

5R

ed

uce

d tim

e [s] V

r= 8

km/s

-32

0

-30

0

-28

0

-26

0

-24

0

-22

0

-20

0

-18

0

-16

0

-14

0

-12

0

-10

0

-80

-60

-40

-20 0

20

40

60

80

Distance [km]

RifS

eis-R

1-N

S

R1

Figure 7

Pg

PsPs

Pg

PiP

PiP

PcPPcP

PmP

PmP

1 1

Ps2

DISTANCE (km)

60

50

40

30

20

10

0

DE

PT

H (

km

)

340 360 380 400 4200 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

a

b

c

R1

S N

Middle Atlas MesetaExternal Rif domain Flysch UnitsInternal Rif domain

AlboranSea

Ele

v. (

km) 2.0

1.5

1.0

0.5

0.0

SassBasin

DISTANCE (km)

14

12

10

86

42

0

T-D

/8 (

s)

340 360 380 400 4200

20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

60

50

40

30

20

10

0

DE

PT

H (

km

)

340 360 380 400 4200 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

01

23

45

67

89

10

111

21

31

41

5R

ed

uce

d tim

e [s] V

r= 8

km/s

-24

0

-22

0

-20

0

-18

0

-16

0

-14

0

-12

0

-10

0

-80

-60

-40

-20 0

20

40

60

80

10

0

12

0

14

0

16

0

Distance [km]

RifSeis-R

2-NS

R2

a

b

c

R2

PcP

Ps2Ps2

Ps1Ps1

PgPg

PiP PcP

PiP

PcPPmP

PmP ?

Figure 8

S N

Middle Atlas MesetaExternal Rif domain Flysch UnitsInternal Rif domain

AlboranSea

Ele

v. (

km) 2.0

1.5

1.0

0.5

0.0

SassBasin

DISTANCE (km)

14

12

10

86

42

0

T-D

/8 (

s)

340 360 380 400 4200

20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

DISTANCE (km)

60

50

40

30

20

10

0

DE

PT

H (

km

)

340 360 380 400 4200 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

Figure 9

a

b

c

01

23

45

67

89

10

1112

13

14

15

Reduce

d tim

e [s] V

r= 8

km/s

-180

-160

-140

-120

-100

-80

-60

-40

-20 0

20

40

60

80

100

120

140

160

180

200

220

240

Distance [km]

RifSeis-R

3-NS

R3

1

2

PmP

Ps

Ps

Pg

PiP

PcP

PmP

Ps

Ps

Pg

PiP

PcP

PmP

1

2R3

S N

Middle Atlas MesetaExternal Rif domain Flysch UnitsInternal Rif domain

AlboranSea

Ele

v. (

km) 2.0

1.5

1.0

0.5

0.0

SassBasin

-2-4-6-8

32

34

36

60

50

40

30

20

10

0

de

pth

(km

)

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420

Distance (km)

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5

velocity (km/s)

Figura10

60

40

20

0

2 4 6 8

depth

km

Vp (km/s)

S N

Middle Atlas Meseta